Open Access
September 2019 Model Criticism in Latent Space
Sohan Seth, Iain Murray, Christopher K. I. Williams
Bayesian Anal. 14(3): 703-725 (September 2019). DOI: 10.1214/18-BA1124


Model criticism is usually carried out by assessing if replicated data generated under the fitted model looks similar to the observed data, see e.g. Gelman, Carlin, Stern, and Rubin (2004, p. 165). This paper presents a method for latent variable models by pulling back the data into the space of latent variables, and carrying out model criticism in that space. Making use of a model's structure enables a more direct assessment of the assumptions made in the prior and likelihood. We demonstrate the method with examples of model criticism in latent space applied to factor analysis, linear dynamical systems and Gaussian processes.


Download Citation

Sohan Seth. Iain Murray. Christopher K. I. Williams. "Model Criticism in Latent Space." Bayesian Anal. 14 (3) 703 - 725, September 2019.


Published: September 2019
First available in Project Euclid: 11 June 2019

zbMATH: 1421.62006
MathSciNet: MR3960767
Digital Object Identifier: 10.1214/18-BA1124

Keywords: factor analysis , Gaussian processes , latent variable models , linear dynamical systems , model criticism

Vol.14 • No. 3 • September 2019
Back to Top